CN107991718B - Mobile phone wearing mode automatic detection method based on multi-mode data analysis - Google Patents

Mobile phone wearing mode automatic detection method based on multi-mode data analysis Download PDF

Info

Publication number
CN107991718B
CN107991718B CN201711211961.7A CN201711211961A CN107991718B CN 107991718 B CN107991718 B CN 107991718B CN 201711211961 A CN201711211961 A CN 201711211961A CN 107991718 B CN107991718 B CN 107991718B
Authority
CN
China
Prior art keywords
mobile phone
data
wearing
value
mode
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201711211961.7A
Other languages
Chinese (zh)
Other versions
CN107991718A (en
Inventor
赵蕴龙
孙龙寿
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN201711211961.7A priority Critical patent/CN107991718B/en
Publication of CN107991718A publication Critical patent/CN107991718A/en
Application granted granted Critical
Publication of CN107991718B publication Critical patent/CN107991718B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V11/00—Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/04—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means
    • G01C21/08—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means involving use of the magnetic field of the earth
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
    • G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
    • G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
    • G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
    • G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
    • G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
    • G01C21/18—Stabilised platforms, e.g. by gyroscope
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Abstract

The invention discloses a mobile phone wearing mode automatic detection method based on multi-mode data analysis, which belongs to the technical field of mobile terminal control and identification, and is characterized in that a mobile phone sensor is used for acquiring distance sensor data and judging whether the distance sensor data is larger than a preset threshold value in a mobile phone to judge whether the mobile phone is in the hand of a wearer; performing data fusion through an accelerometer, a gyroscope and geomagnetic data to obtain a current Euler angle of the mobile phone, and obtaining a current rotation matrix through the Euler angle to judge whether the wearing direction of the mobile phone is correct; and constructing a distance matrix by constructing a section of currently acquired and cut data and sample data of each wearing mode in the mobile phone, taking the minimum value of the optimal matching distances of all the distance matrices, and judging whether the mobile phone is worn on a person or not according to the minimum value. The invention has wider application range and more comprehensive acquired state information, effectively solves the problem of discontinuous direction and is convenient for upper-layer development users to use.

Description

Mobile phone wearing mode automatic detection method based on multi-mode data analysis
Technical Field
The invention relates to a method for automatically detecting a wearing mode of a mobile phone, in particular to a method for automatically detecting a wearing mode of a mobile phone based on multi-mode data analysis, and belongs to the technical field of control and identification of mobile terminals.
Background
With the development of smart phones, applications in all aspects of life are rapidly developed, and particularly, the importance of applications based on smart phones is more and more prominent, such as health monitoring applications based on mobile phones, playing applications based on video and voice information of mobile phones, and the like, which greatly facilitate the life of people in all aspects, but different mobile phone wearing modes have a high influence on the service accuracy prediction precision of the applications.
At present, most of the existing technologies are developed for solving a certain application, and firstly, the existing technologies do not have wide applicability, do not consider various factors of the mobile phone at different levels, different users, different use habits and the like, and also do not consider the requirements of upper-layer applications, and fail to provide a corresponding operation interface, which is convenient for the user to call; secondly, the prior art only gives the wearing position of the mobile phone, sometimes only gives the current angle of the mobile phone based on a geographic coordinate system and the values before and after the conversion of an accelerometer coordinate system, and cannot meet the increasing user requirements; in view of this, the existing operation methods are subject to further improvement.
Disclosure of Invention
The invention mainly aims to provide a mobile phone wearing mode automatic detection method based on multi-mode data analysis, which is used for solving the problems in the prior art.
The purpose of the invention can be achieved by adopting the following technical scheme:
a mobile phone wearing mode automatic detection method based on multi-modal data analysis comprises the following steps:
step S1: starting the process;
step S2: detecting a wearing mode in a hand, acquiring distance sensor data through a mobile phone sensor, and judging whether the distance sensor data is larger than a preset threshold value in the mobile phone to judge whether the mobile phone is in the hand of a wearer;
step S3: detecting the wearing direction of the mobile phone, namely realizing the detection of the wearing direction of the mobile phone through coordinate system conversion, performing data fusion on an accelerometer, a gyroscope and geomagnetic data in sensor data to obtain a current Euler angle of the mobile phone, obtaining a current rotation matrix through the Euler angle to judge whether the wearing direction of the mobile phone is correct, and reminding a wearer of the correct wearing direction;
step S4: detecting the wearing mode of the mobile phone on the person, constructing a distance matrix by a section of data section which is collected and cut at present and sample data of each wearing mode in the mobile phone, taking the minimum value of the optimal matching distances of all the distance matrices, and judging whether the mobile phone is worn on the person according to the minimum value;
step S5: the flow ends.
Further, in step S2, the detecting the wearing style in hand includes the steps of:
step S21: the process flow begins;
step S22: acquiring mobile phone sensor data;
step S23: comparing the distance sensor data with a preset threshold value in the mobile phone, if the distance sensor data is larger than the preset threshold value, considering the wearing mode of the mobile phone as a hand wearing mode, further judging that the mobile phone is in the hand of a wearer, and turning to the step S27;
step S24: the method comprises the steps of obtaining photos of front and rear cameras of a mobile phone, and calculating CDF histogram data of pixel values of each photo;
step S25: respectively taking pixel values of a plurality of quartile points in the two groups of data, calculating the average value of the pixel values, and comparing the average value with a threshold value preset in the mobile phone;
step S26: if the average value is larger than a preset threshold value in the mobile phone, the wearing mode of the mobile phone is considered to be a hand wearing mode, and then the mobile phone is judged to be in the hand of a wearer;
step S27: the flow ends.
Further, in step S3, the detecting of the wearing direction of the mobile phone includes the following steps:
step S31: starting the process;
step S32: converting a coordinate system of data, acquiring data of an accelerometer, a gyroscope and a geomagnetic meter in sensor data, performing data fusion calculation on the accelerometer, the gyroscope and the geomagnetic meter to obtain a current Euler angle of the mobile phone, obtaining a current rotation matrix through the Euler angle, and converting the reading value of the accelerometer based on the mobile phone coordinate system into an acceleration value based on the geodetic coordinate system based on the Euler angle;
step S33: removing noise, namely performing low-pass filtering by using a Butterworth filter, and performing low-pass filtering on noise generated by equipment due to shaking and high-frequency noise generated by a sensor due to a null shift phenomenon to obtain real sensor data;
step S34: data cutting, namely cutting the acquired accelerometer data by using a sliding window algorithm, cutting the filtered accelerometer readings into a plurality of time sequence sections in an isometric mode, wherein 50% of overlapping areas exist between two adjacent data sections, judging whether the wearing direction of the mobile phone is correct or not through the overlapping areas, and reminding a wearer of the correct wearing direction;
step S35: the flow ends.
Further, in step S32, the coordinate system conversion is expressed by equation (1):
Figure GDA0002127658430000031
wherein: rg dA rotation matrix representing a transformation from the device coordinate system d to the geographic coordinate system g;
Vd accrepresenting an accelerometer value based on a mobile phone end;
Vg accrepresenting an accelerometer value based on a geodetic coordinate system.
Further, in step S33, the butterworth low-pass filtering is expressed by equation (2):
Figure GDA0002127658430000032
wherein: w is the angular frequency;
Wcis the cut-off frequency;
n is the order of the filters;
G0is a direct current component.
Further, in step S4, the method for detecting how the mobile phone is worn on the person includes the following steps:
step S41: the process flow begins;
step S42: judging whether the wearing mode detection in the hand is successful, if so, directly jumping out of the body wearing mode detection of the mobile phone, and if not, performing the step S43;
step S43: respectively constructing a distance matrix for the obtained data segment of the accelerometer and the sample data of each wearing mode stored in the mobile phone;
step S44: calculating the optimal matching distance of each distance matrix by using a DTW algorithm;
step S45: taking the minimum value in the optimal matching distances of all the distance matrixes, if the minimum value is smaller than a preset threshold value in the mobile phone, determining that the wearing mode of the current test data is the same as the sample in the distance matrix with the minimum value, and determining that the mobile phone is worn on a person;
step S46: the process flow ends.
Further, in step S4, the matrix terms of the distance matrix are expressed by formula (3):
d(i，j)＝(ai-bj)2(3)
wherein: a isiIs the value of the test sample;
bjis the value of the sample.
The invention has the beneficial technical effects that: according to the method for automatically detecting the wearing mode of the mobile phone based on the multi-mode data analysis, the method for automatically detecting the wearing mode of the mobile phone based on the multi-mode data analysis is wider in application range and more complete in acquired state information, the problem that the reference positions of accelerometers are different in the process of detecting the wearing mode of the mobile phone is solved by using coordinate system conversion, the problem of discontinuous direction is effectively solved, and meanwhile, the method is convenient for upper-layer development users to use more conveniently.
Drawings
FIG. 1 is a general flow chart of a preferred embodiment of a method for automatically detecting a wearing mode of a mobile phone based on multi-modal data analysis according to the present invention;
fig. 2 is a flowchart of a detection method of a wearing manner on a hand of a preferred embodiment of a method for automatically detecting a wearing manner of a mobile phone based on multi-modal data analysis according to the present invention, which may be the same as fig. 1 or different from fig. 1;
fig. 3 is a flow chart of detecting the wearing direction of the mobile phone according to a preferred embodiment of the method for automatically detecting the wearing manner of the mobile phone based on multi-modal data analysis of the present invention, which may be the same as fig. 1 or fig. 2 or different from fig. 1 or fig. 2;
fig. 4 is a flowchart of a wearing manner of a mobile phone according to a preferred embodiment of the method for automatically detecting a wearing manner of a mobile phone based on multi-modal data analysis of the present invention, which may be the same embodiment as fig. 1, 2 or 3, or may be different embodiment from fig. 1, 2 or 3.
Detailed Description
In order to make the technical solutions of the present invention more clear and definite for those skilled in the art, the present invention is further described in detail below with reference to the examples and the accompanying drawings, but the embodiments of the present invention are not limited thereto.
As shown in fig. 1, the method for automatically detecting a wearing manner of a mobile phone based on multi-modal data analysis provided by this embodiment includes the following steps:
step S1: starting the process;
step S2: detecting a wearing mode in a hand, acquiring distance sensor data through a mobile phone sensor, and judging whether the distance sensor data is larger than a preset threshold value in the mobile phone to judge whether the mobile phone is in the hand of a wearer;
step S3: detecting the wearing direction of the mobile phone, namely realizing the detection of the wearing direction of the mobile phone through coordinate system conversion, performing data fusion on an accelerometer, a gyroscope and geomagnetic data in sensor data to obtain a current Euler angle of the mobile phone, obtaining a current rotation matrix through the Euler angle to judge whether the wearing direction of the mobile phone is correct, and reminding a wearer of the correct wearing direction;
step S4: detecting the wearing mode of the mobile phone on the person, constructing a distance matrix by a section of data section which is collected and cut at present and sample data of each wearing mode in the mobile phone, taking the minimum value of the optimal matching distances of all the distance matrices, and judging whether the mobile phone is worn on the person according to the minimum value;
step S5: the flow ends.
Further, in this embodiment, as shown in fig. 2, the step S2 of detecting the wearing style in hand includes the following steps:
step S21: the process flow begins;
step S22: acquiring mobile phone sensor data;
step S23: comparing the distance sensor data with a preset threshold value in the mobile phone, if the distance sensor data is larger than the preset threshold value, considering the wearing mode of the mobile phone as a hand wearing mode, further judging that the mobile phone is in the hand of a wearer, and turning to the step S27;
step S24: the method comprises the steps of obtaining photos of front and rear cameras of a mobile phone, and calculating CDF histogram data of pixel values of each photo;
step S25: respectively taking pixel values of a plurality of quartile points in the two groups of data, calculating the average value of the pixel values, and comparing the average value with a threshold value preset in the mobile phone;
step S26: if the average value is larger than a preset threshold value in the mobile phone, the wearing mode of the mobile phone is considered to be a hand wearing mode, and then the mobile phone is judged to be in the hand of a wearer;
step S27: the flow ends.
Further, in this embodiment, as shown in fig. 3, in the step S3, the mobile phone wearing direction detection includes the following steps:
step S31: starting the process;
step S32: converting a coordinate system of data, acquiring data of an accelerometer, a gyroscope and a geomagnetic meter in sensor data, performing data fusion calculation on the accelerometer, the gyroscope and the geomagnetic meter to obtain a current Euler angle of the mobile phone, obtaining a current rotation matrix through the Euler angle, and converting the reading value of the accelerometer based on the mobile phone coordinate system into an acceleration value based on the geodetic coordinate system based on the Euler angle;
step S33: removing noise, namely performing low-pass filtering by using a Butterworth filter, and performing low-pass filtering on noise generated by equipment due to shaking and high-frequency noise generated by a sensor due to a null shift phenomenon to obtain real sensor data;
step S34: data cutting, namely cutting the acquired accelerometer data by using a sliding window algorithm, cutting the filtered accelerometer readings into a plurality of time sequence sections in an isometric mode, wherein 50% of overlapping areas exist between two adjacent data sections, judging whether the wearing direction of the mobile phone is correct or not through the overlapping areas, and reminding a wearer of the correct wearing direction;
step S35: the flow ends.
Further, in the present embodiment, in the step S32, the coordinate system conversion is expressed by equation (1):
Figure GDA0002127658430000071
wherein: rg dA rotation matrix representing a transformation from the device coordinate system d to the geographic coordinate system g;
Vd accrepresenting an accelerometer value based on a mobile phone end;
Vg accrepresenting an accelerometer value based on a geodetic coordinate system.
Further, in this embodiment, in the step S33, the butterworth low-pass filtering is expressed by equation (2):
Figure GDA0002127658430000072
wherein: w is the angular frequency;
Wcis the cut-off frequency;
n is the order of the filters;
G0is a direct current component.
Further, in this embodiment, as shown in fig. 4, in the step S4, the detecting of the wearing manner of the mobile phone on the person includes the following steps:
step S41: the process flow begins;
step S42: judging whether the wearing mode detection in the hand is successful, if so, directly jumping out of the body wearing mode detection of the mobile phone, and if not, performing the step S43;
step S43: respectively constructing a distance matrix for the obtained data segment of the accelerometer and the sample data of each wearing mode stored in the mobile phone;
step S44: calculating the optimal matching distance of each distance matrix by using a DTW algorithm;
step S45: taking the minimum value in the optimal matching distances of all the distance matrixes, if the minimum value is smaller than a preset threshold value in the mobile phone, determining that the wearing mode of the current test data is the same as the sample in the distance matrix with the minimum value, and determining that the mobile phone is worn on a person;
step S46: the process flow ends.
Further, in this embodiment, in step S4, the matrix items of the distance matrix are expressed by formula (3):
d(i，j)＝(ai-bj)2(3)
wherein: a isiIs the value of the test sample;
bjis the value of the sample.
Example 1:
the embodiment 1 provides a method for collecting samples of various wearing modes in a mobile phone, which includes the following steps:
firstly, a user sets a program into a sample collection mode, selects a position to be collected from position options, and places the mobile phone at a corresponding position, the direction can be free, the mobile phone can normally walk for a distance of more than several minutes (for example, 1 minute), in the walking process, no relative movement between the mobile phone and the body is ensured as much as possible, the wearing mode of the mobile phone is changed, the sample data collection of other wearing modes is carried out again according to the above steps, and the user can completely define and identify the self-defined wearing mode according to the personal requirements.
Example 2:
the embodiment 2 provides an example of automatic identification of a wearing mode of a mobile phone, which mainly includes the following steps:
taking the case that the user places the mobile phone in the trouser pocket as an example, the user firstly initializes the program, then places the mobile phone in the trouser pocket, keeps the mobile phone and the body relatively still, and normally walks for a certain distance. At the moment, the system obtains distance sensor data A, the distance sensor data A is compared with a set distance sensor threshold value B, if A is larger than B, the mobile phone is judged to be in the hand, and if A is smaller than B, subsequent operation is continued; then the system acquires photos according to the front camera and the rear camera to obtain pixel value CDF histogram data of the photos, calculates the mean value C of a plurality of (such as two) percentile points in the photos, compares the mean value C with a set pixel value threshold value D, considers that the mobile phone is in the hand when C is larger than D, and continues to perform subsequent operations when C is smaller than D, i.e. the mobile phone is not in a hand wearing mode; according to the fusion of the current inertial sensor data, the Euler angles phi, theta and psi of the current mobile phone are obtained, the Euler angles can be used for calculating the value (which is constantly changed) of the current mobile phone accelerometer based on the geodetic coordinate system, meanwhile, the time sequence acceleration signal obtained by conversion is cut by a sliding window, and finally, a section of data which is currently collected and cut is acquired, constructing a distance matrix with the sample data segments of each wearing mode in the mobile phone, calculating the optimal matching distance of each distance matrix, and taking the minimum optimal matching distance to obtain the result that the optimal matching distance is the result obtained by forming the distance matrix by the sample data of the trousers pocket and the test data segments, if the user changes the wearing mode of the mobile phone after that, the system gives the predicted values of the wearing position and the wearing direction of the current mobile phone again according to the flow.
Example 3:
if in embodiment 2, the user only places the mobile phone in the hand, the data a obtained by the mobile phone distance sensor will be larger than the set distance sensor data B; if an article or a hand blocks the distance sensor, the data A of the distance sensor is smaller than the threshold B, the processing of the data of the double-camera photos is triggered, the pixel value CDF histogram data of the two camera photos are obtained, a plurality of percentage point data are obtained, the mean value C is calculated and compared with the set pixel value threshold, at least one of the mean value C is larger than the threshold, and the mobile phone is regarded as being in a wearing mode of the hand.
In this embodiment, before the execution, the method may acquire a section of contrast sample data built in the mobile phone, and the user may acquire a wearing mode by walking the mobile phone normally for a period of time (e.g., 1 minute) in a daily wearing mode, and after the wearing mode of the mobile phone is changed, the user may execute the above operations again, in embodiment 1, detailed introduction about sample data acquisition is provided, in the method, the sensor data acquisition period can be set reasonably according to the requirements of different users, each user can conveniently and reasonably utilize the sensor data acquisition period according to the needs of the user, the consumption of the system is reduced, the system can give a calculation result once in each sampling period, meanwhile, the system relates to more parameters, and the system can give each parameter a default value in consideration of the complexity set by the user, but the user can set the corresponding parameter according to the application requirement.
In summary, in this embodiment, according to the method for automatically detecting a wearing manner of a mobile phone based on multimodal data analysis of this embodiment, the method for automatically detecting a wearing manner of a mobile phone based on multimodal data analysis provided by this embodiment has a wider application range and more comprehensive acquired state information, and the problem of different reference positions of accelerometers in the process of detecting a wearing manner of a mobile phone is solved by using coordinate system transformation, so that the problem of discontinuous direction is effectively solved, and meanwhile, the method is convenient for upper-layer development users to use more conveniently.
The above description is only for the purpose of illustrating the present invention and is not intended to limit the scope of the present invention, and any person skilled in the art can substitute or change the technical solution of the present invention and its conception within the scope of the present invention.

Claims (6)

1. A mobile phone wearing mode automatic detection method based on multi-modal data analysis is characterized by comprising the following steps:
step S1: starting the process;
step S2: detecting a wearing mode in a hand, acquiring distance sensor data through a mobile phone sensor, and judging whether the distance sensor data is larger than a preset threshold value in the mobile phone to judge whether the mobile phone is in the hand of a wearer;
step S3: detecting the wearing direction of the mobile phone, namely realizing the detection of the wearing direction of the mobile phone through coordinate system conversion, performing data fusion on an accelerometer, a gyroscope and geomagnetic data in sensor data to obtain a current Euler angle of the mobile phone, obtaining a current rotation matrix through the Euler angle to judge whether the wearing direction of the mobile phone is correct, and reminding a wearer of the correct wearing direction, wherein in the step S3, the detection of the wearing direction of the mobile phone comprises the following steps: step S31: starting the process; step S32: converting a coordinate system of data, acquiring data of an accelerometer, a gyroscope and a geomagnetic meter in sensor data, performing data fusion calculation on the accelerometer, the gyroscope and the geomagnetic meter to obtain a current Euler angle of the mobile phone, obtaining a current rotation matrix through the Euler angle, and converting the reading value of the accelerometer based on the mobile phone coordinate system into an acceleration value based on the geodetic coordinate system based on the Euler angle; step S33: removing noise, namely performing low-pass filtering on the acceleration value under the geodetic coordinate system by using a Butterworth filter, and performing low-pass filtering on noise generated by equipment due to shaking and high-frequency noise generated by a sensor due to a null shift phenomenon to obtain a real acceleration value; step S34: data cutting, namely cutting the filtered acceleration value by using a sliding window algorithm, cutting the filtered acceleration value into a plurality of time sequence data sections in an equal length mode, wherein 50% of overlapping areas exist between two adjacent data sections, judging whether the wearing direction of the mobile phone is correct or not through the overlapping areas, and reminding a wearer of the correct wearing direction; step S35: the flow is finished;
step S4: detecting the wearing mode of the mobile phone on the human body, constructing a distance matrix by using the obtained data segment of the accelerometer and sample data of each wearing mode in the mobile phone, taking the minimum value of the optimal matching distances of all the distance matrices, and judging whether the mobile phone is worn on the human body according to the minimum value;
step S5: the flow ends.
2. The method for automatically detecting the wearing manner of the mobile phone based on the multi-modal data analysis as claimed in claim 1, wherein the step S2 of detecting the wearing manner in the hand comprises the following steps:
step S21: the process flow begins;
step S22: acquiring mobile phone sensor data;
step S23: comparing the distance sensor data with a preset threshold value in the mobile phone, if the distance sensor data is larger than the preset threshold value, considering the wearing mode of the mobile phone as a hand wearing mode, further judging that the mobile phone is in the hand of a wearer, and turning to the step S27;
step S24: the method comprises the steps of obtaining photos of front and rear cameras of a mobile phone, and calculating CDF histogram data of pixel values of each photo;
step S25: respectively taking pixel values of a plurality of quartile points in the two groups of data, calculating the average value of the pixel values, and comparing the average value with a threshold value preset in the mobile phone;
step S26: if the average value is larger than a preset threshold value in the mobile phone, the wearing mode of the mobile phone is considered to be a hand wearing mode, and then the mobile phone is judged to be in the hand of a wearer;
step S27: the flow ends.
3. The method for automatically detecting the wearing style of a mobile phone based on multi-modal data analysis of claim 1, wherein in step S32, the coordinate system transformation is expressed by formula (1):
Figure FDA0002385684740000021
wherein: rg dA rotation matrix representing a transformation from the device coordinate system d to the geographic coordinate system g;
Vd accrepresenting the reading value of an accelerometer based on a mobile phone end;
Vg accrepresenting acceleration values based on a geodetic coordinate system.
4. The method for automatically detecting the wearing style of a mobile phone based on multi-modal data analysis as claimed in claim 1, wherein in step S33, the butterworth filter low-pass filtering is expressed by formula (2):
Figure FDA0002385684740000022
wherein: w is the angular frequency;
Wcis the cut-off frequency;
n is the order of the filters;
G0is a direct current component.
5. The method for automatically detecting the wearing mode of the mobile phone based on the multi-modal data analysis as claimed in claim 1, wherein the step S4 of detecting the wearing mode of the mobile phone on the person comprises the following steps:
step S41: the process flow begins;
step S42: judging whether the wearing mode detection in the hand is successful, if so, directly jumping out of the body wearing mode detection of the mobile phone, and if not, performing the step S43;
step S43: respectively constructing a distance matrix for the obtained data segment of the accelerometer and the sample data of each wearing mode stored in the mobile phone;
step S44: calculating the optimal matching distance of each distance matrix by using a DTW algorithm;
step S45: taking the minimum value in the optimal matching distances of all the distance matrixes, if the minimum value is smaller than a preset threshold value in the mobile phone, determining that the wearing mode of the current test data is the same as the sample in the distance matrix with the minimum value, and determining that the mobile phone is worn on a person;
step S46: the process flow ends.
6. The method for automatically detecting the wearing style of a mobile phone based on multi-modal data analysis of claim 5, wherein in step S4, the matrix term of the distance matrix is represented by formula (3):
d(i，j)＝(ai-bj)2(3)
wherein: a isiIs the value of the test sample;
bjis the value of the sample.
CN201711211961.7A 2017-11-28 2017-11-28 Mobile phone wearing mode automatic detection method based on multi-mode data analysis Active CN107991718B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711211961.7A CN107991718B (en) 2017-11-28 2017-11-28 Mobile phone wearing mode automatic detection method based on multi-mode data analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711211961.7A CN107991718B (en) 2017-11-28 2017-11-28 Mobile phone wearing mode automatic detection method based on multi-mode data analysis

Publications (2)

Publication Number Publication Date
CN107991718A CN107991718A (en) 2018-05-04
CN107991718B true CN107991718B (en) 2020-06-30

Family

ID=62033460

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711211961.7A Active CN107991718B (en) 2017-11-28 2017-11-28 Mobile phone wearing mode automatic detection method based on multi-mode data analysis

Country Status (1)

Country Link
CN (1) CN107991718B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108917756A (en) * 2018-06-12 2018-11-30 四川斐讯信息技术有限公司 A kind of course estimation method and system
CN110517463B (en) * 2019-08-16 2021-11-02 维沃移动通信有限公司 Method for reminding user to wear wearable device and mobile terminal
CN111248922B (en) * 2020-02-11 2022-05-17 中国科学院半导体研究所 Human body respiration condition acquisition paste based on accelerometer and gyroscope and preparation method thereof

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103076619A (en) * 2012-12-27 2013-05-01 山东大学 System and method for performing indoor and outdoor 3D (Three-Dimensional) seamless positioning and gesture measuring on fire man
CN203588195U (en) * 2013-12-04 2014-05-07 温慎洁 Wearable computing device with adjacency distance sensor
CN105942678A (en) * 2016-07-18 2016-09-21 广东乐源数字技术有限公司 Method for judging whether bracelet is not worn on hand
CN106249581A (en) * 2016-07-28 2016-12-21 努比亚技术有限公司 A kind of intelligent watch, the charging method of intelligent watch and device
CN107085231A (en) * 2017-04-24 2017-08-22 济南中景电子科技有限公司 The detection method of location call equipment and its wearing mode with adaptation function

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8092398B2 (en) * 2005-08-09 2012-01-10 Massachusetts Eye & Ear Infirmary Multi-axis tilt estimation and fall remediation
US20150145670A1 (en) * 2013-11-26 2015-05-28 Kabushiki Kaisha Toshiba Electronic device and control method thereof

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103076619A (en) * 2012-12-27 2013-05-01 山东大学 System and method for performing indoor and outdoor 3D (Three-Dimensional) seamless positioning and gesture measuring on fire man
CN203588195U (en) * 2013-12-04 2014-05-07 温慎洁 Wearable computing device with adjacency distance sensor
CN105942678A (en) * 2016-07-18 2016-09-21 广东乐源数字技术有限公司 Method for judging whether bracelet is not worn on hand
CN106249581A (en) * 2016-07-28 2016-12-21 努比亚技术有限公司 A kind of intelligent watch, the charging method of intelligent watch and device
CN107085231A (en) * 2017-04-24 2017-08-22 济南中景电子科技有限公司 The detection method of location call equipment and its wearing mode with adaptation function

Also Published As

Publication number Publication date
CN107991718A (en) 2018-05-04

Similar Documents

Publication Publication Date Title
CN108596976B (en) Method, device and equipment for relocating camera attitude tracking process and storage medium
CN107991718B (en) Mobile phone wearing mode automatic detection method based on multi-mode data analysis
EP3162284A1 (en) Communication method, apparatus and system for wearable device
US10142598B2 (en) Wearable terminal device, photographing system, and photographing method
CN108633249B (en) Physiological signal quality judgment method and device
CN106357898A (en) Method and system for reminding user not to walk and simultaneously play with mobile phones on basis of intelligent wearable equipment
CN105847581B (en) User reminding method and user terminal
CN106510164B (en) Luggage case control method and device
CN113892920B (en) Wearing detection method and device of wearable equipment and electronic equipment
CN108418867A (en) A kind of method of environmental monitoring preventing human interference
CN114510542B (en) Method for generating motion trail, electronic equipment and server
US20220060666A1 (en) Wearable device for performing detection of events by utilizing camera module and wireless communication device
CN108139564A (en) Focusing control apparatus, photographic device, focusing control method and focusing control program
CN110006445B (en) Running distance calculation method and device
CN106248101A (en) A kind of monitoring method and device
JP2015170257A (en) Input method, program and input device
CN111189469A (en) Step counting method, terminal device and storage medium
CN110717399A (en) Face recognition method and electronic terminal equipment
CN106527429B (en) Luggage case control method and device
US20150220613A1 (en) Relationship estimation device and relationship estimation method
CN108564059B (en) Wearable device, data processing method and device thereof, equipment and storage medium
CN111803902B (en) Swimming stroke identification method and device, wearable device and storage medium
WO2018014432A1 (en) Voice application triggering control method, device and terminal
CN111026273A (en) Automatic setting method and device for intelligent wearable equipment, electronic equipment and storage medium
CN111623799A (en) Terminal state detection method and device, terminal and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant